Senior GPU LLM Training Performance Architect

NVIDIA Gruppe; Santa Clara (CA); On-site; USD 184,000 - 287,500; Full time

Benefits offered by this job

Equity

Comprehensive benefits

Job summary

NVIDIA Gruppe is looking for an expert to optimize AI training workloads on cutting-edge hardware and software platforms. Applicants should have a PhD in a relevant field, extensive experience in deep learning, and strong programming skills in C++, Python, and CUDA.

The position offers a competitive salary range of $184,000 to $287,500 for Level 4 employees, and $224,000 to $356,500 for Level 5, along with equity and benefits.

Qualifications

• 5+ years of experience with a PhD, or 8+ years with an MS.
• Strong background in deep learning and neural networks.
• Deep understanding of computer architecture and GPU fundamentals.
• Experience analyzing and tuning performance at multiple levels.

Responsibilities

• Analyze, profile, and optimize AI training workloads.
• Implement software across NVIDIA's deep learning platform.
• Support NVIDIA submissions to MLPerf Training benchmarks.
• Build tools for workload analysis and optimization.

Skills

Deep learning

Neural networks

Programming in C++

Programming in Python

CUDA programming

Education

PhD in Computer Science, Electrical Engineering or Computer Engineering

MS or equivalent experience

Tools

NVIDIA deep learning platform stack

Job description

NVIDIA Gruppe is looking for an expert to optimize AI training workloads on cutting-edge hardware and software platforms. Applicants should have a PhD in a relevant field, extensive experience in deep learning, and strong programming skills in C++, Python, and CUDA.

The position offers a competitive salary range of $184,000 to $287,500 for Level 4 employees, and $224,000 to $356,500 for Level 5, along with equity and benefits.